Our studies on the metabolism and DNA adduction of the heterocyclic amines (HAs) have shown that these compounds are procarcinogens that require metabolic activation for genotoxicity. These compounds, produced in meats during the process of cooking, are mutagens in the Ames salmonella assay upon metabolic activation by cytochrome P450. Carcinogenicity and metabolism studies of three HAs have been undertaken in cynomolgus monkeys. We found that 2-amino-3-methyl-imidazo[4,5- f]quinoline (IQ) is a potent hepatocarcinogen in cynomolgus monkeys. We determined that metabolic processing by cytochrome P450-mediated N- hydroxylation and phase II esterification is a critical feature for HA carcinogenicity in nonhuman primates. Four phase II pathways have been shown to contribute to the activation of the N-hydroxylamines of IQ, 2- amino-3,8-dimethylimidazo[4,5-f]-quinoxaline (MeIQx), and 2-amino-1- methyl-6-phenylimidazo[4,5-b]-pyridine (PhIP): O-acetyltransferase, sulfotransferase, t-RNA synthetase, and phosphorylase. The major DNA adduct of IQ, MeIQx, and PhIP has been identified as the respective C8- guanine adduct, indicating a similarity among these compounds in DNA adducts formed. Studies with PhIP show that cynomolgus monkeys possess a good capacity to metabolically activate this compound in vivo and that this activation results in DNA adduction that can be found in all tissues examined to date, including white blood cells. From these metabolism studies, we predict that PhIP will be carcinogenic to nonhuman primates. In contrast, the metabolic activation of MeIQx is low in the cynomolgus monkeys and correlates with a lack of tumorigenicity or longer latency period (studies ongoing). We have examined the metabolism of HAs in the mammary gland of rats and determined that the O-acetyl-transferase is the principal phase II enzyme in the mammary tissue. PhIP metabolites in breast milk have been identified. The role that various mammary metabolites play in mammary carcinogenesis is currently being assessed.